Develop

.github/workflows/python-CD.yml

@@ -12,7 +12,7 @@ on:
     branches: [master]
 jobs:
   format_and_check:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-20.04
     environment: production
     steps:
       - uses: actions/checkout@v3
@@ -30,7 +30,7 @@ jobs:
 
   build:
     needs: format_and_check
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-20.04
     environment: production
     strategy:
       matrix:
@@ -62,7 +62,7 @@ jobs:
 
   deploy:
     needs: build
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-20.04
     environment: production
     steps:
       - uses: actions/checkout@v3
@@ -97,7 +97,7 @@ jobs:
 
   release:
     needs: deploy
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-20.04
     environment: production
     steps:
       - uses: actions/checkout@v3
@@ -128,7 +128,7 @@ jobs:
 
   deploy_docs:
     needs: build
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-20.04
     environment: production
     steps:
       - uses: actions/checkout@v3

.github/workflows/python-CI.yml

@@ -11,7 +11,7 @@ on:
     branches: [develop, feature/*]
 jobs:
   format_and_check:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-20.04
     environment: development
     steps:
       - uses: actions/checkout@v3
@@ -29,7 +29,7 @@ jobs:
 
   build:
     needs: format_and_check
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-20.04
     environment: development
     strategy:
       matrix:

.python-version

@@ -0,0 +1 @@
+3.10.16

Pipfile

@@ -4,21 +4,13 @@ verify_ssl = true
 name = "pypi"
 
 [packages]
-tqdm = "*"
-numpy = "*"
-scikit-image = "*"
-splitraster = {editable = true, path = "."}
+numpy = "<2.0.0,>=1.19.0"
+tqdm = "<5.0.0,>=4.40.0"
+scikit-image = "<1.0.0,>=0.18.0"
+pytest = "*"
 
 [dev-packages]
 ipykernel = "*"
-mkdocs = "*"
-black = "*"
-pre-commit = "*"
-pytest = "*"
-ipykernel = "*"
-torch = "*"
-matplotlib = "*"
-torchvision = "*"
 
 [requires]
 python_version = "3.10"

notebooks/Tutorial_II.ipynb

@@ -85,7 +85,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/Users/caihaocui/.local/share/virtualenvs/split_raster-m2hVat3N/lib/python3.10/site-packages/osgeo/gdal.py:312: FutureWarning: Neither gdal.UseExceptions() nor gdal.DontUseExceptions() has been explicitly called. In GDAL 4.0, exceptions will be enabled by default.\n",
+      "/Users/caihaocui/.local/share/virtualenvs/split_raster-co_bDcoB/lib/python3.10/site-packages/osgeo/gdal.py:314: FutureWarning: Neither gdal.UseExceptions() nor gdal.DontUseExceptions() has been explicitly called. In GDAL 4.0, exceptions will be enabled by default.\n",
       "  warnings.warn(\n"
      ]
     },
@@ -102,7 +102,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Generating: 100%|\u001b[32m██████████\u001b[0m| 400/400 [00:00<00:00, 2127.55img/s]\n"
+      "Generating: 100%|\u001b[32m██████████\u001b[0m| 400/400 [00:00<00:00, 1114.16img/s]\n"
      ]
     },
     {
@@ -119,7 +119,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Generating: 100%|\u001b[32m██████████\u001b[0m| 400/400 [00:00<00:00, 2673.49img/s]"
+      "Generating: 100%|\u001b[32m██████████\u001b[0m| 400/400 [00:00<00:00, 2581.73img/s]"
      ]
     },
     {
@@ -256,7 +256,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Generating: 100%|\u001b[32m██████████\u001b[0m| 20/20 [00:00<00:00, 692.43img/s]"
+      "Generating: 100%|\u001b[32m██████████\u001b[0m| 20/20 [00:00<00:00, 227.96img/s]"
      ]
     },
     {
@@ -308,14 +308,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Latest run time 2024-03-23 16:59:37\n"
+      "Latest run time 2025-03-23 20:37:15\n"
      ]
     }
    ],
@@ -334,11 +334,9 @@
   }
  ],
  "metadata": {
-  "interpreter": {
-   "hash": "67d32777f905cb443c7595cdac206bb322335f5bfbf8ae9544bce83f7472f717"
-  },
   "kernelspec": {
-   "display_name": "Python 3.8.10 64-bit ('base': conda)",
+   "display_name": "split_raster-co_bDcoB",
+   "language": "python",
    "name": "python3"
   },
   "language_info": {
@@ -351,7 +349,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.13"
+   "version": "3.10.16"
   },
   "orig_nbformat": 4
  },

requirements.txt

@@ -2,3 +2,4 @@ numpy>=1.19.0, <2.0.0
 tqdm>=4.40.0, <5.0.0
 scikit-image>=0.18.0, <1.0.0
 # GDAL==3.8.4 # For GIS only on MacOS `brew install gdal` and `pip install GDAL==3.8.4`
+# python_version >= "3.10"

setup.py

@@ -10,7 +10,7 @@ def read_requirements():
 
 setup(
     name="splitraster",
-    version="0.3.5",
+    version="0.3.6",
     author="Chris Cui",
     license="MIT",
     platforms="any",

src/splitraster/geo.py

@@ -1,14 +1,12 @@
 from tqdm import tqdm
-
-from osgeo import gdal
-from osgeo import gdal_array
-
+from osgeo import gdal, gdal_array
 import numpy as np
 from pathlib import Path
 import random
+from typing import Tuple, Optional
 
 
-def read_rasterArray(image_path):
+def read_rasterArray(image_path: str) -> Tuple[np.ndarray, Tuple[float, ...], str]:
     dataset = gdal.Open(image_path, gdal.GA_ReadOnly)
     image = dataset.ReadAsArray()  # get the rasterArray
     # convert 2D raster to [1, H, W] format
@@ -19,7 +17,9 @@ def read_rasterArray(image_path):
     return image, geotrans, proj
 
 
-def save_rasterGeoTIF(im_data, im_geotrans, im_proj, file_name):
+def save_rasterGeoTIF(
+    im_data: np.ndarray, im_geotrans: Tuple[float, ...], im_proj: str, file_name: str
+) -> None:
     if Path(file_name).is_file():
         print(f"Overwrite existing file: {file_name}")
 
@@ -40,41 +40,28 @@ def save_rasterGeoTIF(im_data, im_geotrans, im_proj, file_name):
     dataset = driver.Create(
         file_name, int(im_width), int(im_height), int(im_bands), datatype
     )
-    if dataset != None:
+    if dataset is not None:
         dataset.SetGeoTransform(im_geotrans)
         dataset.SetProjection(im_proj)
     for i in range(im_bands):
         dataset.GetRasterBand(i + 1).WriteArray(im_data[i])
     del dataset
 
 
-# def save_rasterArray(im_data, path, image_prototype_path):
-
-#     output = gdal_array.SaveArray(
-#         im_data, path, format="GTiff",  prototype=image_prototype_path)
-#     return True
-
-
-def save_rasterArray(im_data, file_name):
+def save_rasterArray(im_data: np.ndarray, file_name: str) -> bool:
     if Path(file_name).is_file():
         print(f"Overwrite existing file: {file_name}")
-    output = gdal_array.SaveArray(im_data, file_name, format="GTiff")
+    gdal_array.SaveArray(im_data, file_name, format="GTiff")
     return True
 
 
-def count_files(folder_path):
-    count = 0
-    for path in Path(folder_path).iterdir():
-        if path.is_file():
-            count += 1
-    return count
+def count_files(folder_path: str) -> int:
+    return sum(1 for path in Path(folder_path).iterdir() if path.is_file())
 
 
-def padding_mul_image(img, stride):
-    D = img.shape[0]  # this one is for (D, H, W) format Channel First.
-    height = img.shape[1]
-    width = img.shape[2]
-    # get the minial padding image size
+def padding_mul_image(img: np.ndarray, stride: int) -> np.ndarray:
+    D, height, width = img.shape  # (D, H, W) format Channel First.
+    # get the minimal padding image size
     H = int(np.ceil(height / stride) * stride)
     W = int(np.ceil(width / stride) * stride)
 
@@ -84,13 +71,17 @@ def padding_mul_image(img, stride):
         padded_img[d, :, :] = np.pad(
             onelayer, ((0, H - height), (0, W - width)), "reflect"
         )
-    # padded_img = np.squeeze(padded_img)  # Remove axes of length one
     return padded_img
 
 
 def split_image(
-    img_path, save_path, crop_size, repetition_rate=0, overwrite=True, ext="."
-):
+    img_path: str,
+    save_path: str,
+    crop_size: int,
+    repetition_rate: float = 0,
+    overwrite: bool = True,
+    ext: Optional[str] = ".",
+) -> Optional[int]:
     # check input image
     img, geotrans, proj = read_rasterArray(img_path)
     if img is None:
@@ -104,7 +95,7 @@ def split_image(
     print(f"Input Image File Shape (D, H, W):{ img.shape}")
 
     stride = int(crop_size * (1 - repetition_rate))
-    print(f"{crop_size=}, {stride=}")
+    print(f"crop_size = {crop_size}, stride = {stride}")
 
     padded_img = padding_mul_image(img, stride)
 
@@ -138,32 +129,36 @@ def tile_generator():
             crop_img = padded_img[:, h : h + crop_size, w : w + crop_size]
             crop_image_name = f"{new_name:04d}{ext}"
             crop_image_path = Path(save_path) / crop_image_name
-            # save_rasterArray(crop_img,  str(crop_image_path)) # just save the raster image
             save_rasterGeoTIF(crop_img, geotrans, proj, str(crop_image_path))
-            new_name = new_name + 1
+            new_name += 1
             pbar.update(1)
 
     return n + 1
 
 
 def random_crop_image(
-    img_path,
-    img_save_path,
-    label_path,
-    label_save_path,
-    crop_size=256,
-    crop_number=20,
-    img_ext=".tif",
-    label_ext=".tif",
-    overwrite=True,
-):
+    img_path: str,
+    img_save_path: str,
+    label_path: str,
+    label_save_path: str,
+    crop_size: int = 256,
+    crop_number: int = 20,
+    img_ext: str = ".tif",
+    label_ext: str = ".tif",
+    overwrite: bool = True,
+) -> Optional[int]:
     """Generate Random cropped image pair from the input image pairs.
 
     Args:
         img_path (str): path of input image
-        img_save_path (str):
+        img_save_path (str): path to save cropped images
+        label_path (str): path of input label
+        label_save_path (str): path to save cropped labels
         crop_size (int): image tile size (H,W), i.e., 256x256
-        overwrite (bool, optional): [overwrite existing files]. Defaults to True.
+        crop_number (int): number of crops to generate
+        img_ext (str): extension for image files
+        label_ext (str): extension for label files
+        overwrite (bool): overwrite existing files
     """
     img, geotrans, proj = read_rasterArray(label_path)
     if img is None:
@@ -231,8 +226,8 @@ def random_crop_image(
             crop_image_path = Path(label_save_path) / crop_image_name
             save_rasterGeoTIF(labelCrop, geotrans, proj, str(crop_image_path))
 
-            new_name = new_name + 1  # update image name
-            crop_cnt = crop_cnt + 1  # add crop count
+            new_name += 1  # update image name
+            crop_cnt += 1  # add crop count
             pbar.update(1)
 
     return crop_cnt  # return total crop sample pair number.